Bug 383551

Summary: hpet_late_init panic on VIA Apollo CLE266
Product: [Fedora] Fedora Reporter: Warren Togami <wtogami>
Component: kernelAssignee: Thomas Gleixner <tglx>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: low Docs Contact:
Priority: low    
Version: 8CC: kernel-maint
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2007-11-28 20:23:45 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 188611    
Attachments:
Description Flags
lspci -v none

Description Warren Togami 2007-11-14 22:30:54 UTC
kernel-2.6.23.1-42.fc8.i586
kernel-2.6.23.1-49.fc8.i586

http://www.disklessworkstations.com/cgi-bin/web/200029.html
Intermittently (10-20%) when booting this VIA based thin client, panics very
early in hpet_late_init.  nohpet seems to avoid the problem.  Other
distributions are reportedly not seeing this issue, so this might be a
relatively new kernel bug.

Comment 1 Warren Togami 2007-11-14 22:31:12 UTC
http://people.redhat.com/wtogami/temp/via-i586-hpet-panic.jpg
Photo of panic

Comment 2 Warren Togami 2007-11-14 22:32:04 UTC
Created attachment 259061 [details]
lspci -v

Comment 3 Warren Togami 2007-11-14 22:37:34 UTC
davej suggested assigning this to tglx.


Comment 4 Chuck Ebbert 2007-11-14 23:43:34 UTC
The code dump at the bottom of the page doesn't match the i586 kernel at all.

8d b4 26 00 00 00 00  is a NOP [leal 0(esi),esi]



Comment 5 Thomas Gleixner 2007-11-15 00:21:24 UTC
> davej suggested assigning this to tglx.

/me grumbles :)

Warren,

"Intermittently (10-20%) when booting this VIA based thin client, panics very
early in hpet_late_init."

Can you please provide the full boot log from one of the other (80-90%)
successful boots ?

Further it might be helpful, if you could capture the boot of a crash via serial
console.

Please add
 earlyprintk=serial,ttyS0,115200

to the kernel command line and connect the box to a second one with a null
modem cable. On the second box please capture the output of the kernel.

Thanks,

    tglx


Comment 6 Warren Togami 2007-11-15 18:22:18 UTC
I will have physical access to it again tomorrow.


Comment 7 Chuck Ebbert 2007-11-28 20:23:45 UTC
More info is available in the newer bug, including full oops captures.
Why a NOP instruction could cause a divide error is a real mystery...

*** This bug has been marked as a duplicate of 394981 ***